DESCRIPTION (Applicant's Description) A conference entitled "Multi disciplinary Conference on Causality and Statistics" will be held at the University of California, Berkeley, August 8-11, 2000. The conference is jointly sponsored by the Bernoulli Society and the Econometrics Laboratory of UC Berkeley as well as the applicants of this application. The impetus for this conference is the development in the 1980's and l990's of a unified mathematical and statistical model of causation. The model unifies the approach developed in the statistical and epidemiologic literature with that developed in the sociology, philosophy and econometrics literature. The goals of this conference are to: (I) bring leading researchers, junior investigators, and graduate students from each of the above disciplines together to share methodological insights and discoveries among disciplines; (ii) teach researchers how to translate their discipline-specific causal language into the new formalism; (iii) produce an edited survey book on causal inference based on the conference, including an overview article on the relevance for biomedical statistics; (iv) promote the topic to the press and the public on the last half-day of the conference with an open session on the vital role of statistics and causal inference in policy initiatives. We expect about 60 participants supported by this NIH application, and about 35 econometricians supported by the Econometrics Laboratory National Science Foundation grant. Plenary sessions consist of invited talks with two invited discussants. Didactic sessions on two mornings give an overview. Evenings include open microphone sessions and contributed poster sessions. The final afternoon is devoted to the press and the public.